| # ============================================================================ | |
| # Prep: Run full data build → data/output | |
| # ============================================================================ | |
| # Working directory must be the project root (folder containing data/ and Rscripts/). | |
| # | |
| # Order (fast steps first; heavy raster build last): | |
| # 1. GBIF DuckDB → sf-gbif.parquet → gbif_census_ndvi_anno.parquet | |
| # 2. CalEnviroScreen + SF EJ → GeoPackages | |
| # 3. GTFS zip + timetable + stop headways (~20–30 s; cached for app startup) | |
| # 4. Greenspace distance rasters (DuckDB; slow) | |
| # 5. RSF Program nearest rasters (DuckDB; same grid as 4) | |
| # 6. Per-CBG greenspace coverage CSV (precomputed so the app's first isochrone | |
| # doesn't recompute the CBG x greenspace intersection) | |
| # | |
| # Then upload files from data/output/ to the HuggingFace dataset (manually or | |
| # see comments in upload_to_huggingface.R). | |
| source("Rscripts/prep/create_annotated_gbif_parquet.R") | |
| source("Rscripts/prep/build_equity_layers.R") | |
| source("Rscripts/prep/implement_optimizations.R") | |
| source("Rscripts/prep/making-greenspace-raster.R") | |
| source("Rscripts/prep/making-rsfprogram-raster.R") | |
| source("Rscripts/prep/build_cbg_greenspace_coverage.R") | |